[🐛 Bug]: postgresql connection tune

[Viste]

Describe the bug
Now you can't tune connection from app side

To Reproduce
run app with postgress tuned idle connection live settings and get:

{"worker_type": "arqworker", "asctime": "2025-12-09 14:09:45,850", "message": "Exception during reset or similar", "levelname": "ERROR", "name": "sqlalchemy.pool.impl.QueuePool", "filename": "base.py", "otelTraceID": "4bc123179dd4dc92086a4d9333467cd6", "otelSpanID": "66e372f202b7142c", "otelTraceSampled": true, "otelServiceName": "keep-api", "threadName": "process_event_worker_1", "process": 34, "module": "base", "exc_info": "Traceback (most recent call last):\n  File \"/venv/lib/python3.13/site-packages/sqlalchemy/pool/base.py\", line 986, in _finalize_fairy\n    fairy._reset(\n    ~~~~~~~~~~~~^\n        pool,\n        ^^^^^\n    ...<2 lines>...\n        asyncio_safe=can_manipulate_connection,\n        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n    )\n    ^\n  File \"/venv/lib/python3.13/site-packages/sqlalchemy/pool/base.py\", line 1432, in _reset\n    pool._dialect.do_rollback(self)\n    ~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^\n  File \"/venv/lib/python3.13/site-packages/sqlalchemy/engine/default.py\", line 699, in do_rollback\n    dbapi_connection.rollback()\n    ~~~~~~~~~~~~~~~~~~~~~~~~~^^\npsycopg2.OperationalError: server closed the connection unexpectedly\n\tThis probably means the server terminated abnormally\n\tbefore or while processing the request.\n", "taskName": null}

Expected behavior
you can tune connection settings
Screenshots
If applicable, add screenshots to help explain your problem.

Additional context
Add any other context about the problem here.

[dosubot]

You can tune PostgreSQL connection pool settings in KeepHQ from the application side using environment variables: DATABASE_POOL_SIZE (default: 5), DATABASE_MAX_OVERFLOW (default: 10), and KEEP_DB_PRE_PING_ENABLED for health checks. These are used when the SQLAlchemy engine is created, so you can set them in your deployment environment to adjust pool size and overflow limits as needed see code see docs.

Other pool parameters like pool_recycle or pool_timeout are not currently exposed for tuning; adding support for those would require code changes in keep/api/core/db_utils.py.

The error psycopg2.OperationalError: server closed the connection unexpectedly is likely related to idle connections being terminated by the PostgreSQL server or a mismatch between server-side timeouts and the pool settings. If you need to tune idle connection timeouts or recycling, you may need to extend the codebase to expose those parameters.

Let me know if you need details on how to extend the configuration for more advanced tuning.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}